ResearchTasksPolicy Question AnsweringPolicy Question AnsweringFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedPolicyBenchDeepseek-R164.34Accuracy11Apr 15, 2026PolicyBench USDeepseek-R166.43Accuracy11Apr 15, 2026PolicyBench ChineseQwQ-32B65.33Accuracy11Apr 15, 2026